Towards Black-box Iterative Machine Teaching
This addresses the challenge of efficient teaching in scenarios with limited observability, offering a method for faster convergence in machine learning systems, though it appears incremental as it builds on existing teaching frameworks.
The paper tackles the problem of black-box machine teaching where teacher and learner use different feature representations, proposing an active teacher model that queries the learner to estimate its status and provably achieves faster convergence than passive learning, with sample complexities provided and experimental verification against an omniscient teacher.
In this paper, we make an important step towards the black-box machine teaching by considering the cross-space machine teaching, where the teacher and the learner use different feature representations and the teacher can not fully observe the learner's model. In such scenario, we study how the teacher is still able to teach the learner to achieve faster convergence rate than the traditional passive learning. We propose an active teacher model that can actively query the learner (i.e., make the learner take exams) for estimating the learner's status and provably guide the learner to achieve faster convergence. The sample complexities for both teaching and query are provided. In the experiments, we compare the proposed active teacher with the omniscient teacher and verify the effectiveness of the active teacher model.